The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
# TrumpSignal 📊
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
> Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## What it does
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Ingests Trump's Truth Social posts daily from HuggingFace
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Predicts whether the next trading day will be high or low market impact
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Shows real-time stock price movements around each post
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Provides semantic search over all posts by topic
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Live Demo
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
https://huggingface.co/spaces/Ailee52/trump-signal
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Pipeline
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Step | Component | Description |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
|---|---|---|
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 2 | SQLite Database | Local storage, DVC tracked |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 4 | XGBoost Classifier | Predicts next-day market impact |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 5 | FastAPI | Serves predictions and search via API |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| 6 | Streamlit | Interactive frontend on port 7860 |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Tech Stack
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Layer | Technology |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
|---|---|
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Data | HuggingFace Datasets, SQLite, DVC |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| ML | XGBoost, scikit-learn, sentence-transformers |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| API | FastAPI, uvicorn |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Frontend | Streamlit |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Deployment | Docker, HuggingFace Spaces |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
| Scheduling | APScheduler, GitHub Actions |
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Run Locally
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
Requirements: Docker
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
```bash
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
# Clone
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
git clone https://github.com/Rogersurf/trump-signal
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
cd trump-signal
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
# Build and run
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
docker build -t trump-signal .
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
docker run -p 8000:8000 -p 7860:7860 trump-signal
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
```
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Frontend: http://localhost:7860
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- API docs: http://localhost:8000/docs
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Run without Docker
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
```bash
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
pip install -r requirements.txt
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
pip install -e .
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
# Terminal 1 — API
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
# Terminal 2 — Frontend
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
python -m streamlit run frontend/streamlitapp.py
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
```
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Project Structure
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
trump-signal/
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── app/api/ # FastAPI endpoints
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── backend/ # ML training and inference
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── backend_database/ # Data ingestion and SQLite
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── frontend/ # Streamlit pages
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── tests/ # Unit tests
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
├── Dockerfile # Container setup
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
└── requirements.txt
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Dataset
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
## Team
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Rogerio Braunschweiger De Freitas Lima
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Chenhao Lou
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Suchanya Baiyam (Ailee)
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
MSc Economics and Business Administration (Business Data Science)
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
Aalborg University, 2026
title: TrumpSignal emoji: 📊 colorFrom: red colorTo: blue sdk: docker pinned: false
TrumpSignal 📊
An MLOps pipeline that ingests Trump's Truth Social posts and generates market impact signals.
Disclaimer: This project was developed solely for academic purposes as part of the Data Engineering and Machine Learning Operations in Business course at Aalborg University. It is not affiliated with, endorsed by, or intended to influence any political party, candidate, or movement. The analysis is purely technical and should not be interpreted as political commentary or financial advice.
What it does
- Ingests Trump's Truth Social posts daily from HuggingFace
- Classifies posts by category (threatening, self-promotion, attacking, etc.)
- Predicts whether the next trading day will be high or low market impact
- Shows real-time stock price movements around each post
- Provides semantic search over all posts by topic
Live Demo
https://huggingface.co/spaces/Ailee52/trump-signal
Pipeline
| Step | Component | Description |
|---|---|---|
| 1 | HuggingFace Dataset | Source of Trump's Truth Social posts |
| 2 | SQLite Database | Local storage, DVC tracked |
| 3 | Embeddings Cache | MiniLM-L6-v2 vectors for semantic search |
| 4 | XGBoost Classifier | Predicts next-day market impact |
| 5 | FastAPI | Serves predictions and search via API |
| 6 | Streamlit | Interactive frontend on port 7860 |
Tech Stack
| Layer | Technology |
|---|---|
| Data | HuggingFace Datasets, SQLite, DVC |
| ML | XGBoost, scikit-learn, sentence-transformers |
| API | FastAPI, uvicorn |
| Frontend | Streamlit |
| Deployment | Docker, HuggingFace Spaces |
| Scheduling | APScheduler, GitHub Actions |
Run Locally
Requirements: Docker
# Clone
git clone https://github.com/Rogersurf/trump-signal
cd trump-signal
# Build and run
docker build -t trump-signal .
docker run -p 8000:8000 -p 7860:7860 trump-signal
- Frontend: http://localhost:7860
- API docs: http://localhost:8000/docs
Run without Docker
pip install -r requirements.txt
pip install -e .
# Terminal 1 — API
python -m uvicorn app.api.main:app --host 0.0.0.0 --port 8000
# Terminal 2 — Frontend
python -m streamlit run frontend/streamlitapp.py
Project Structure
trump-signal/ ├── app/api/ # FastAPI endpoints ├── backend/ # ML training and inference ├── backend_database/ # Data ingestion and SQLite ├── frontend/ # Streamlit pages ├── tests/ # Unit tests ├── Dockerfile # Container setup └── requirements.txt
Dataset
chrissoria/trump-truth-social — updated daily, includes post categories, engagement metrics, stock price snapshots, and GDELT global event indicators.
Team
- Rogerio Braunschweiger De Freitas Lima
- Chenhao Lou
- Suchanya Baiyam (Ailee)
MSc Economics and Business Administration (Business Data Science)
Aalborg University, 2026
- Downloads last month
- 104